The broad objective of this proposal is to identify combinatorial vaccine approaches that will result in effective immunization against mesothelin-expressing pancreatic adenocarcinomas. In a completed phase I trial, two allogeneic GM-CSF-secreting pancreatic cell lines demonstrated a dose-dependent induction of T cell immunity in patients with resected disease. Analysis of post-vaccination immune responses identified mesothelin as a candidate new target against which hoth T cell and antibody responses were directed in patients who remain disease-free. However, recent findings strongly suggest that immune tolerance mechanisms provide a formidable barrier to effective immunization in most cancer patients. In this proposal, we will develop and optimize a mesothelin-directed vaccine, identify the most therapeutically potent combinations of vaccine and immune modulating agents, and explore the synergistic interaction between these treatment modalities. Understanding the mechanisms of interaction should lead to the design of rational combinations with improved efficacy against pancreatic adenocarcinoma. Five specific aims are proposed. Aim 1 will identify and characterize mesothelin-specific tumor vaccines that can induce immune responses potent enough to treat mesothelin-expressing Panc02 tumors. A GM-CSF secreting, mesothelin-expressing whole tumor cell approach and a Listeria Monocytogenes (LM) containing mesothelin vaccine approach, will be evaluated. Mesothelin-specific DNA vaccines developed in project 2 will also be tested in this model. Aim 2 will optimize T regulatory cell (Treg) inhibiting agents that enhance anti-tumor immunity when combined with the mesothelin-targeted vaccines. Two existing and one new Treg inhibiting agent (project 3) will be evaluated in the Panc02 model. Aim 3 will optimize combinations of Treg depleting agents that enhance mesothelin-targeted vaccine potency when also given with the antibody to the anti-CTLA-4 T cell regulatory pathway, to enhance T cell activation. In aim 4, the immune modulating agent/vaccine combinations that are effective in the less stringent transplantable tumor model will be tested for their ability to treat naturally developing Panc02 hepatic metastases following hemisplenic transplantation. Aim 5 will compare the most potent vaccine/anti-CTLA-4/and Treg cell-inhibiting agent combinations for enhanced efficacy when combined with new agents targeted at other immune regulatory checkpoints (projects 3 and 4). These combinatorial approaches will first be tested against transplantable Panc02, and then against natural Panc02 metastases. The final outcome will be the identification of a mesothelin-targeted, combinatorial vaccine approach for testing in patients.